Kidney transplants fail over time. Injury due to antibodies that target cells of the blood vessels within the transplant kidney (microcirculatory endothelium) is a major reason why the transplants fail. Clinical tools to predict this injury are not available. MicroRNAs in the blood could serve as markers of injury. We propose to develop microRNA (miRNA)-based blood tests for the diagnosis of antibody-mediated microcirculatory inflammation/injury. We have carefully selected a group of 180 kidney transplant recipients at our hospital who had serum samples collected and stored in our laboratory at the time of undergoing an allograft biopsy. The AMI group (n=60) consists of 60 serum samples from 60 kidney transplant recipients with biopsy diagnosis of antibody-mediated microcirculatory inflammation. The No AMI group (n=120) consists of 120 serum samples from 120 kidney transplant recipients whose biopsy did not have such inflammation. This group consists of patients with biopsy diagnosis of acute cellular rejection, acute tubular injury, and normal allograft biopsy (n=40 each). The Specific Aims are to: (1) Identify circulating extracellular miRNAs in serum of kidney graft recipients that are diagnostic of AMI in the kidney allograft, (2) Develop a statistical model (signature) diagnostic of AMI in the kidney allograft, (3) Validate the diagnostic model in serum samples obtained from an independent external cohort of kidney transplant recipients. We will use high throughput barcoded deep sequencing to characterize the serum miRNA transcriptome. We will use miRNA-specific qPCR assay to accurately quantify selected individual miRNAs, for their quick translation to the clinic. We will compare the serum miRNA transcriptome from the two groups (AMI vs. No AMI) and identify miRNAs that are different between the two. We will then develop PCR assays to accurately quantify the selected miRNAs. We will develop statistical models consisting of combination of miRNAs that best predict the diagnosis. We will statistically determine the clinical benefit of using such a model fr diagnosing AMI. To ensure that the statistical model works in different groups of patients, we will do PCR assays and test the statistical model (signature) in another group of 105 kidney transplant recipients (35 AMI and 70 No AMI) from a different hospital.